#include "./opr_impl.h"
#include "./algos.h"
#include "src/common/algo_chooser.h"
#include "src/common/utils.cuh"
#include "src/fallback/handle.h"
using namespace megdnn;
using namespace fallback;
std::vector<BatchedMatrixMulForwardImpl::Algorithm*> BatchedMatrixMulForwardImpl::
get_all_algorithms(
const TensorLayout& A, const TensorLayout& B, const TensorLayout& C) {
AlgoBase::SizeArgs args{this, A, B, C};
return megdnn::get_all_algorithms<BatchedMatrixMulForwardImpl>(args);
}
std::vector<BatchedMatrixMulForwardImpl::Algorithm*> BatchedMatrixMulForwardImpl::
get_all_algorithms_safe(
const TensorLayout& A, const TensorLayout& B, const TensorLayout& C) {
AlgoBase::SizeArgs args{this, A, B, C};
return megdnn::get_all_algorithms_safe<BatchedMatrixMulForwardImpl>(args);
}
BatchedMatrixMulForwardImpl::Algorithm* BatchedMatrixMulForwardImpl::
get_algorithm_heuristic(
const TensorLayout& A, const TensorLayout& B, const TensorLayout& C,
size_t workspace_limit_in_bytes, const AlgoAttribute& positive_attr,
const AlgoAttribute& negative_attr) {
AlgoBase::SizeArgs args{this, A, B, C};
if (sm_algo_pack.algo_default.is_available_attribute(
args, positive_attr, negative_attr, workspace_limit_in_bytes)) {
return &sm_algo_pack.algo_default;
}
return megdnn::get_algo_match_attribute<BatchedMatrixMulForwardImpl>(
sm_algo_pack.all_algos, args, workspace_limit_in_bytes,
"batched matrix mul forward", positive_attr, negative_attr);
}
size_t BatchedMatrixMulForwardImpl::get_workspace_in_bytes(
const TensorLayout& A, const TensorLayout& B, const TensorLayout& C) {
return get_dnn_workspace(this, A, B, C);
}
void BatchedMatrixMulForwardImpl::exec(
_megdnn_tensor_in A, _megdnn_tensor_in B, _megdnn_tensor_out C,
_megdnn_workspace workspace) {
check_exec(A.layout, B.layout, C.layout, workspace.size);
AlgoBase::ExecArgs args(this, A, B, C, workspace);
auto&& algo = get_algorithm(this, A.layout, B.layout, C.layout);
algo->exec(args);
}